Original Article

Craniofacial growth of Class III subjects six to sixteen years of age

Sara M. Wolfea; Eustaquio Araujob; Rolf G. Behrentsc; Peter H. Buschangd

ABSTRACT
Objective: To characterize the mixed-longitudinal craniofacial growth of untreated, white, Class III
subjects 6 to 16 years of age.
Materials and Methods: Serial cephalograms of 19 females and 23 males with Class III
malocclusion were evaluated at three time points (6–8, 10–12, and 14–16 years of age). A similar
number of Class I controls were randomly selected and matched for age and sex. The
cephalograms were traced and digitized, and 20 variables were evaluated. Growth patterns were
quantified, and class and sex differences were evaluated using multi-level analyses.
Results: In comparison with Class I subjects, Class III subjects had significantly (P # .05) larger
mandibular plane angles, gonial angles, mandibular ramus heights, mandibular corpus lengths,
and SNB angles, with differences that were maintained between 6 and 16 years of age. Maxillary
lengths and ANB angles were significantly smaller and remained smaller in Class III subjects than
in Class I subjects. Lower face height, maxillary-mandibular differential, and mandibular body
length were also significantly larger and increased significantly more between 6 and 16 years of
age in Class III subjects. The WITS appraisal was significantly smaller in Class III subjects and
decreased significantly more over time. Most linear measures showed significant sex differences
favoring males; the angular measures and anteroposterior (AP) maxillomandibular relationships
showed no sex differences.
Conclusions: The AP maxillomandibular relationship of Class III subjects worsens over time. AP
discrepancies are primarily due to excessive mandibular growth, which produces a protrusive,
hyperdivergent phenotype. The AP discrepancies of males are larger than those of females, with
differences increasing over time. (Angle Orthod. 2011;81:211–216.)
KEY WORDS: Class III; Cephalometrics; Whites; Growth

INTRODUCTION more in extreme cases.’’ Regardless of the definition

used, the orthodontist’s understanding of how untreat-
Since 1737, when Bourdet first described the
ed Class III whites grow has been limited by the low
skeletal pattern of children with protruding chins, Class
prevalence of Class III malocclusions and the tenden-
III malocclusions have been characterized in various
cy to treat subjects at a younger age. The prevalence
ways. Angle1 defined Class III malocclusion as ‘‘the
of Class III malocclusion has been reported2,3 to range
relation of the jaws with all the lower teeth occluding
between 1.6% and 12.2%. National health surveys4,5
mesial to normal the width of one premolar or even
have shown that 4.9% of white children between 6 and
11 years of age and 6% of youths between 12 and
Private Practice, Birmingham, AL.
a
17 years of age display a bilateral mesiocclusion.
Professor, Orthodontic Department, St Louis University, St
b
Using negative overjet to classify Class III subjects,
Louis, Mo.
c
Professor and Department Chair, Orthodontic Department,
more recent surveys6,7 indicate a 1–4% prevalence of
St Louis University, St Louis, Mo. Class III malocclusion among North Americans.
d
Professor, Department of Orthodontics, Baylor College of Previous cross-sectional cephalometric character-
Dentistry, Dallas, Tex. izations have shown that when compared to Class I
Corresponding author: Dr Peter Buschang, Department of whites, Class III subjects have substantially smaller
Orthodontics, Baylor College of Dentistry, 3302 Gaston Ave,
Dallas, TX 75246 ANB angles,8–11 slightly smaller SNA angles,8,10,12,13
(e-mail: PHBuschang@bcd.tamhsc.edu) and substantially larger SNB angles.8,10,11,13 The saddle
Accepted: June 2010. Submitted: April 2010.
and cranial base flexure angles have also been shown
G 2011 by The EH Angle Education and Research Foundation, to be more acute among Class III subjects;9,14–16 a lack
Inc. of differences in cranial base angles has also been

DOI: 10.2319/051010-252.1 211 Angle Orthodontist, Vol 81, No 2, 2011

212
reported.8,11,13 Similarly, MacDonald and coworkers16
found no significant differences between Class III and
Class I malocclusions for either the SNA angle or the
maxillary depth. Larger mandibular dimensions have
commonly been cited8,10,13 as the predominant charac-
teristic in the Class III subject.
While serial data are required to determine actual
growth differences associated with untreated Class III
malocclusions, relatively few longitudinal studies have
been conducted. Baccetti et al.,17 who evaluated 22
white Class III subjects at two time points, showed that
maxillomandibular relationships worsen over time.
However, because of the lack of untreated controls,
they were not able to determine growth deficiencies or
excesses of the craniofacial components. Alexander et
al.,18 who described the longitudinal growth between 4
and 20 years of age of 103 Class III whites, also
showed definite worsening of anteroposterior (AP)
skeletal relationships, but they were also unable to
characterize the differences, again as a result of the
lack of controls.
In order to better understand the development of
skeletal differences among Class III subjects, this study
was designed to evaluate the growth of Class III and
matched Class I subjects between 6 and 16 years of age.
MATERIALS AND METHODS
Serial cephalometric radiographs were selected from
the Bolton-Brush Growth Study Center in Cleveland,
Ohio; these radiographs represented white children
aged 6–8, 10–12, and 14–16 years of age. Forty-two
Class III subjects were selected based on their molar
relationships, as determined by the Bolton-Brush
Growth Study. Forty age-group and sex-matched Class
I controls were randomly selected from the same
sample. The subjects were classified as Class III or
Class I during the early permanent dentition, based on
clinical observations and dental models. Subjects with
cleft lip, cleft palate, and other craniofacial syndromes
were excluded. The sample was mixed longitudinal; all
of the subjects had records at two of the three age
groupings; 34% of the sample had complete longitudinal
series comprising three records.
Cephalometric Tracing and Analysis
Lateral head films were traced on 0.003-inch frosted
acetate. Each film was traced by one investigator and
checked for accuracy by one of two investigators. Ten
percent of the films were randomly chosen and
retraced to assess reliability. All films were then
digitized with a Numonics Accugrid Digitizer (Numo-
nics Corp, Montgomeryville, Penn) and analyzed with
the Dentofacial Planner software program, version
7.0.2 (Toronto, Ontario, Canada). Sixteen landmarks
Angle Orthodontist, Vol 81, No 2, 2011
WOLFE, ARAUJO, BEHRENTS, BUSCHANG
were identified and digitized. The magnification (ap-
proximately 6%) was not corrected.
Cephalometric measures were derived from the
analyses of Jarabak and Fizzel,19 Jacobson,20 McNa-
mara,21 and Steiner22; they represent a variety of AP
and vertical measurements reported to be significant in
Class III development.
Statistical Analysis
Data were analyzed using multi-level statistical
models.23 Multi-level statistical analysis does not make
the assumption of complete longitudinal data, nor does
it require exact intervals between age groups, making
it well suited for this mixed-longitudinal study. Growth
curves were described as polynomials and estimated
using iterative generalized least squares. The regres-
sions consisted of intercept (size) and age (growth
velocity at 11 years of age) terms. To center the
intercept, 11 was subtracted from the subjects’ ages
(ie, ages 7, 11, and 15 were changes to 24, 0, and 4,
respectively). Separate analyses were performed to
evaluate class and sex differences.
Each multi-level model estimated the constant and
age terms, as well as group differences in the constant
and age terms. The models’ constant terms described
the size or the angle of either Class I malocclusions or
females, depending on the analysis, at 11 years of
age. The age terms described the yearly growth
changes. The multi-level models also estimated group
differences (Class III minus Class I; male minus
female) for both the constant and age terms.
Reliability analysis was performed using the Dahl-
berg method’s error statistic [! (Sdeviations2/2n)].
The method errors of the linear measures ranged
between 0.74 mm and 2.1 mm, with mid-face length
(Co-A) showing the greatest error. Angular measure-
ment method errors ranged between 0.8u and 2.9u,
with the cranial base angle (Ba-S-N) showing the
greatest error.
RESULTS
The multi-level models showed significant growth
changes for all of the variables except the cranial base
angle (Ba-S-N), articular angle (S-Ar-Go), and WITS
(Table 1). Eleven of the 20 measures (55%) showed
statistically significant differences between Class I and
Class III malocclusions. Lower face height (ANS-Me),
corpus length (Go-Pg), and the maxillomandibular
differential (Mx-Md) were significantly larger in 11
year-old Class III subjects and demonstrated signifi-
cantly greater growth increases over time than Class I
subjects (Figure 1). The WITS appraisal was signifi-
cantly smaller at 11 years and it decreased signifi-
cantly more between 6–16 years in Class Ill’s than in
GROWTH OF CLASS III SUBJECTS 213

Table 1. Multi-level Growth Estimates for Untreated Class I Subjects and Class Differences (Class III 2 Class I) Between 6 and 16 Years of Age
Class I Subjects Class III 2 Class I Differences
Intercept Age Effects Intercept Age Effects
At 11 y of Standard Change Standard Standard Differences in Standard
Age Error Over Time Error Differences Error Change Over Time Error
N-Me* 112.58 0.58 2.27 0.07 — — — —
ANS-Me* 63.06 0.72 1.08 0.08 1.64 1.01 0.22 0.10
N-ANS* 50.08 0.28 1.01 0.04 — — — —
N-ANS/ANS-Me* 78.66 0.73 0.18 0.08 — — — —
MPA* 30.84 0.84 20.31 0.05 2.44 1.14 — —
S-N* 68.37 0.33 0.80 0.03 — — — —
Ba-S-N 130.10 0.56 20.08 0.07 — — — —
N-S-Ar* 123.12 0.55 0.17 0.08 — — — —
S-Ar-Go 139.20 0.60 20.07 0.10 — — — —
Ar-Go-Me* 131.97 0.75 20.38 0.05 2.23 1.02 — —
Co-A* 84.09 0.43 1.34 0.06 — — — —
ANS-PNS* 51.17 0.39 0.73 0.05 21.55 0.53 — —
Ar-Go* 44.81 0.45 1.18 0.06 1.37 0.59 — —
Go-Pg* 66.56 0.69 1.56 0.08 2.07 0.97 0.34 0.10
Co-Gn* 108.25 0.75 2.60 0.07 3.91 1.04 — —
ANB* 3.24 0.36 20.25 0.03 22.29 0.49 — —
SNA* 80.28 0.46 0.10 0.04 — — — —
SNB* 76.92 0.58 0.37 0.04 2.46 0.82 — —
Mx-Md* 23.50 0.60 1.11 0.09 5.29 0.85 0.25 0.12
WITS* 20.81 0.35 0.02 0.08 23.93 0.49 20.27 0.11
* Significant (P , .05) growth changes; — 5 Not statistically significant; Mx-Md 5 Co-Gn minus Co-A; Wits 5 A ) functional occlusal plane
minus B ) functional occlusal plane.

Class l’s. The mandibular plane angle (MPA), gonial length (ANS-PNS) and ANB angle were significantly
angle (Ar-Go-Me), ramus height (Ar-Go), mandibular smaller in Class III malocclusions, Figure 2.
length (Co-Gn), and the SNB angle were all significantly The multi-level models showed statistically significant
larger in the Class III group at 11 years of age; maxillary sex differences for eight measures (Table 2). The gonial

Figure 1. Measures demonstrating significant size differences at 11 years and significant growth differences between Class I subjects and
Class III subjects 6–16 years of age.

Angle Orthodontist, Vol 81, No 2, 2011

214 WOLFE, ARAUJO, BEHRENTS, BUSCHANG

Figure 2. Measures demonstrating significant size differences at 11 years of age between Class I subjects and Class III subjects.

angle (Ar-Go-Me) was significantly greater in males than
in females at age 11, with no significant growth
differences. Total face height (N-Me), lower face height
(ANS-Me), upper face height (N-ANS), anterior cranial
base length (S-N), mid-face length (Co-A), mandibular
body length (Go-Pg), and mandibular length (Co-Gn)
were all significantly larger in males and demonstrated
significantly greater growth increases over time com-
pared with females. Mandibular ramus height (Ar-Go)
was significantly smaller in 11-year-old males but
showed significantly greater increases for males than for
females.

Angle Orthodontist, Vol 81, No 2, 2011

GROWTH OF CLASS III SUBJECTS 215

Table 2. Multi-level Growth Estimates of Female Growth Changes and Sex Differences (male 2 females) Between 6 and 16 Years of Age
Female Male 2 Female Differences
Intercept Age Effects Intercept Age Effects
At 11 y of Standard Change Over Standard Standard Differences in Standard
Age Error Time Error Differences Error Change Over Time Error
N-Me* 111.11 0.81 1.95 0.06 2.98 1.13 0.64 0.12
ANS-Me* 62.56 0.70 1.02 0.07 2.66 0.99 0.35 0.10
N-ANS* 49.81 0.41 0.86 0.06 0.54 0.57 0.28 0.08
N-ANS/ANS-Me* 78.66 0.73 0.18 0.08 — — — —
MPA* 32.08 0.62 20.31 0.05 — — — —
S-N* 67.95 0.46 0.70 0.04 0.83 0.65 0.20 0.06
Ba-S-N 130.11 0.56 20.08 0.06 — — — —
N-S-Ar* 123.12 0.55 0.17 0.08 — — — —
S-Ar-Go 139.20 0.60 20.07 0.10 — — — —
Ar-Go-Me* 133.13 0.57 20.40 0.05 — — — —
Co-A* 83.13 0.60 1.20 0.09 1.91 0.84 0.30 0.12
ANS-PNS* 50.39 0.29 0.73 0.06 — — — —
Ar-Go* 45.59 0.49 1.03 0.08 20.12 0.69 0.29 0.12
Go-Pg* 66.56 0.69 1.56 0.08 2.07 0.97 0.34 0.10
Co-Gn* 109.13 0.82 2.34 0.09 2.27 1.15 0.53 0.13
ANB* 2.09 0.29 20.25 0.03 — — — —
SNA* 80.28 0.46 0.10 0.04 — — — —
SNB* 78.17 0.43 0.37 0.04 — — — —
Mx-Md* 26.17 0.52 1.24 0.06 — — — —
WITS* 22.79 0.33 20.13 0.06 — — — —
* Significant (P , .05) growth changes; — 5 Not statistically significant; Mx-Md 5 Co-Gn minus Co-A; Wits 5 A H functional occlusal plane
minus B H functional occlusal plane.

DISCUSSION In contrast to those of Class I subjects, the

AP relationships of Class III subjects clearly worsen
between 6 and 16 years of age. Compared to Class I
subjects, Class III subjects had smaller ANB and larger
SNB angles, as previously reported9,11,13 for samples
evaluated cross sectionally. As expected, the Class III
subjects in the present study also exhibited a
significantly larger maxillomandibular differential and
a smaller WITS differential. Importantly, both of these
differentials worsened over time, indicating a worsen-
ing of the Class III malocclusion. Decreases in the
WITS measures and increases in the maxillomandib-
ular differential have been previously reported17 for
Class III subjects followed longitudinally. The ANB
angle in the present study probably did not worsen
over time as a result of the greater-than-expected
increases in lower facial height exhibited by Class III
subjects. Similar increases in lower face height of
Class III subjects have been reported18 between 4 and
20 years in age. It is possible that the ANB angle
maintained the same growth rates in Class I subjects
and Class III subjects because the anterior movements
of point B were masked by the inferior movements of
point B. The WITS better represents the true AP
changes because it is measured from the occlusal
plane and is unaffected by the vertical changes that
occur.
mandibles of Class III subjects were more hyperdiver-
gent and substantially larger. The angular differences
identified among Class III subjects in the present
study, including the increased mandibular plane and
larger gonial angles, have been previously well
established.8,10,12 Supporting the present findings,
greater ramus heights have been reported13 for Class
III than for Class I subjects. Total mandibular length
has also been previously shown8,9 to be significantly
larger in Class III subjects of similar ages. Increased
corpus length among Class III subjects compared with
Class I subjects has been previously identified by
Jacobson and coworkers.10 The greater growth in-
creases in corpus length identified in the present study
have not been previously shown. This indicates that it
is the remodeling pattern (ie, a development of a more
obtuse gonial angle and increases in corpus length
associated with deposition of bone at the lower
posterior aspect of the ramus), rather than condylar
growth, that is the primary determinant of overall
mandibular excess among Class III subjects.
In contrast to their large, prognathic mandibles, the
maxillas of the Class III subjects in the present study
were orthognathic. MacDonald et al.16 also showed no
significant differences between Class III subjects and
Class I subjects for either SNA or maxillary depth.
While most other cross-sectional studies8,10,12,13 have

Angle Orthodontist, Vol 81, No 2, 2011

216
reported maxillary retrusion among Class III subjects,
their results tend to be limited and inconsistent. For
example, Guyer et al.8 reported significant differences
in the SNA angle between Class I subjects and Class
III subjects for three of the four age groups evaluated;
Battagel11 only found differences after all of the group
data had been combined; Tollaro et al.13 did not find
significant differences among their 4- and 5-year-old
subsamples, but they did report differences for the 6-
year-olds and for the entire sample combined. Taken
together, the present and previous studies indicate that
even though the maxillas of Class III subjects are
smaller, maxillary retrusion is relatively mild and
represents only a minor contribution to the develop-
ment of AP discrepancies.
Sex differences, which increased over time, were
evident for most of the linear measures. Males were
larger than females, and the differences increased with
age. These results are consistent with an understand-
ing of craniofacial and somatic growth. Sex differences
in maxillary and mandibular growth favoring males
have been previously established.24,25 Sex differences
are small during childhood and become pronounced
during adolescence, as a result of the two extra years
of childhood growth among males as well as the
greater intensity of the male adolescent spurt.
CONCLUSIONS
N Maxillomandibular relationships of Class III subjects
progressively worsen between 6 and 16 years of
age.
N Class III subjects have a somewhat smaller, but not
more recessive, maxilla than do Class I subjects;
maxillary size differences are established early and
maintained through 16 years of age.
N Class III subjects have larger, more protrusive
mandibles than Class I subjects, with AP growth
excesses that accumulate over time. Class III
subjects also have hyperdivergent mandibles and
excessive growth of lower facial height.
N Males are larger than females, with size differences
increasing between 6 and 16 years of age. There
were no significant sex differences in AP maxillo-
mandibular and angular relationships.
REFERENCES
1. Angle EH. Classification of malocclusion. Dent Cosmos.
1899;41:248.
2. Huber RE, Reynolds JW. A dentofacial study of male
students at the University of Michigan in the physical
hardening program. Am J Orthod. 1946;32:1–21.
3. Ast DB, Carlos JP, Cons NC. The prevalence and
characteristics of malocclusion among senior high school
students in upstate New York. Am J Orthod. 1965;51:
437–445.
Angle Orthodontist, Vol 81, No 2, 2011
WOLFE, ARAUJO, BEHRENTS, BUSCHANG
4. Kelly JE, Sanchez M, Van Kirk LE. An assessment of the
occlusion of teeth of children 6–11 years, United States.
Washington, DC: US DHEW Publication (HRA)74-1612; 1973.
5. Kelly JE, Harvey CJ. An assessment of the occlusion of the
teeth in youths 12–17 years, United States. Washington,
DC: US DHEW Publication (HRA)77-1644; 1977.
6. Proffit W, Fields H, Moray L. Prevalence of malocclusion
and orthodontic treatment need in the United States:
estimates from the NHANES III survey. Int J Adult Orthod
Orthognath Surg. 1998;13:97–106.
7. Mills LF. Epidemiologic studies of occlusion. IV. The
prevalence of malocclusion in a population of 1455 school
children. J Dent Res. 1966;45:332–336.
8. Guyer EC, Ellis EE, McNamara JA, Behrents RG. Compo-
nents of Class III malocclusion in juveniles and adolescents.
Angle Orthod. 1986;56:7–30.
9. Reyes BC, Baccetti T, McNamara JA. An estimate of
craniofacial growth in Class III malocclusion. Angle Orthod.
2006;76:577–584.
10. Jacobson A, Evans WG, Preston CB, Sadowsky PL.
Mandibular prognathism. Am J Orthod. 1974;66:140–171.
11. Battagel J. The aetiological factors in Class III malocclusion.
Eur J Orthod. 1993;15:347–370.
12. Sanborn RT. Differences between the facial skeletal
patterns of Class III malocclusion and normal occlusion.
Angle Orthod. 1955;25:208–222.
13. Tollaro I, Baccetti T, Bassarelli V, Franchi L. Class III
malocclusion in the deciduous dentition, a morphological
and correlation study. Eur J Orthod. 1994;16:401–408.
14. Proff P, Will F, Bokan I, Fanghanel J, Gedrange T. Cranial
base features in skeletal Class III patients. Angle Orthod.
2008;78:433–439.
15. Jarvinen S. Saddle angle and maxillary prognathism: a
radiological analysis of the association between the NSAr
and SNA angles. Br J Orthod. 1984;11:209–213.
16. MacDonald KE, Kapust AJ, Turley PK. Cephalometric
changes after the correction of Class III malocclusion with
maxillary expansion/facemask therapy. Am J Orthod Den-
tofacial Orthop. 1999;116:13–24.
17. Baccetti T, Franchi L, McNamara JA Jr. Growth in the
untreated Class III subject. Semin Orthod. 2007;13:130–
142.
18. Alexander AE, McNamara JA Jr, Franchi L, Baccetti T.
Semilongitudinal cephalometric study of craniofacial growth
in untreated Class III malocclusion. Am J Orthod Dentofacial
Orthop. 2009;135:700.e1–14.
19. Jarabak J, Fizzel J. Technique and Treatment with Light
Wire Edgewise Appliances. St Louis, Mo: Mosby; 1972.
20. Jacobson A. The ‘‘Wits’’ appraisal of jaw disharmony.
Am J Orthod. 1975;67:125–138.
21. McNamara JA. A method of cephalometric evaluation.
Am J Orthod. 1984;86:449–469.
22. Steiner C. Cephalometrics for you and me. Am J Orthod.
1953;39:729–755.
23. Hoeksma JB, van der Beek MCJ. Multilevel modelling of
longitudinal cephalometric data explained for orthodontists.
Eur J Orthod. 1991;13:197–201.
24. Buschang PH, Tanguay R, Demirjian A, LaPalme L, Gold-
stein H. Sexual dimorphism in mandibular growth of French-
Canadian children 6 to 10 years of age. Am J Phys
Anthropol. 1986;71:33–37.
25. Riolo ML, Moyers RE, McNamara JA Jr, Hunter WS. An
Atlas of Craniofacial Growth. Monograph 2, Center for
Human Growth and Development, University of Michigan.
Ann Arbor, Mich: University of Michigan; 1974.